Interdisciplinary Thesis Project
The Interdisciplinary Thesis Project is a cornerstone of the M.Tech in Data Science program at IIT Guwahati. Spanning the second year of the program, this intensive research component enables students to apply advanced data science methodologies to complex, real-world problems across multiple domains. Jointly offered by the Mehta Family School of Data Science and Artificial Intelligence, the Department of Mathematics, and the Department of Electronics and Electrical Engineering, the thesis project encourages cross-disciplinary collaboration and innovation. Students work closely with faculty members from diverse academic backgrounds, involving co-supervision across departments.
Key Highlights
- Duration 1 full academic year
- Collaboration Students are encouraged to collaborate with research labs, industry partners, or interdisciplinary faculty teams.
- Scope Projects span domains such as healthcare analytics, computer vision, natural language processing, finance, IoT, robotics, and smart energy systems, among others.
- Outcome Many thesis outcomes lead to publications in top-tier conferences or journals, open-source contributions, and solutions with real-world impact.
Objectives
Problem-Solving with Data
Equip students with the skills to formulate and solve complex, real-world problems using advanced data science techniques. The focus is on developing practical, scalable solutions grounded in data-driven decision-making.
Interdisciplinary Integration
Encourage students to synthesize concepts from computer science, mathematics, statistics, and domain-specific knowledge. This cross-disciplinary approach is key to addressing multifaceted challenges in modern data science applications.
Research and Critical Thinking
Cultivate a strong foundation in research methodology and analytical reasoning. Students are trained to approach problems with curiosity and rigor, enabling them to contribute to cutting-edge innovations in data science.
Guidance and Evaluation
Each student is assigned a thesis supervisor or co-supervisors, depending on the project’s scope. Evaluation is carried out through regular progress reviews, a mid-term seminar, and a final thesis defense before a faculty panel. This project prepares students for careers in R&D, academia, or data-driven leadership roles in industry, offering a rigorous foundation for innovation at the intersection of data science and real-world challenges.
Recent Thesis Projects
Below are some thesis titles from our recent graduated batch, showcasing the breadth of research areas and applications our students engage in.
| Recurrent Neural Network for a Micro Gas Sensor System |
| Face Recognition and Gait Analysis using AI-Enabled Techniques |
| Generative Models in Pathology |
| Sentiment Analysis for Social Networks using LLMs |
| Identifying Communities in Online Social Networks using Deep Learning Techniques |
| Multi-task Learning for Medical Image Generation and Segmentation |
| Multi-modal Deepfake Detection |
| News Classification for Stock Price Prediction with Tiny LLMs |
| Deep Learning Method for Photoacoustic Image Reconstruction in Medical Imaging |
| Optimizing NLP Tasks Through Data Augmentation with LLMs |
| Leveraging Deep Learning for the Inverse Design of Nanophotonic Devices |
| Deep Learning-based UAV Trajectory Optimization in Beyond 5G Communication Systems |
| AI and Portfolio Theory-Based Resource Allocation and Energy Efficiency in 6G Wireless Communication Networks |
| Dynamics of Neural Networks in Fractals |